public inbox for dev@dpdk.org
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: Junlong Wang <wang.junlong1@zte.com.cn>
Cc: dev@dpdk.org
Subject: Re: [PATCH v3 0/5] net/zxdh: add support some new features
Date: Sat, 31 Jan 2026 10:11:57 -0800	[thread overview]
Message-ID: <20260131101157.4700d0e9@phoenix.local> (raw)
In-Reply-To: <20260131062208.2445121-1-wang.junlong1@zte.com.cn>

On Sat, 31 Jan 2026 14:22:03 +0800
Junlong Wang <wang.junlong1@zte.com.cn> wrote:

> V3:
>   - modify some issues from AI review.
> 
> V2:
>   - add updated release note and the documentation.
>   - remove unnecessary initialization.
>   - using the compiler builtins(__rte_clz32).
>   - remove C++ style comments.
> 
> V1:
>   - support modifying queue depth.
>   - optimeze alloc queue resources.
>   - support set link speed.
>   - support primary/secondary process.
>   - support GENEVE TSO/chksum,and tunnel packets outer udp chksum.
> 
> Junlong Wang (5):
>   net/zxdh: add support for modifying queue depth
>   net/zxdh: optimize alloc queue resources
>   net/zxdh: add link speed set and autoneg get support
>   net/zxdh: add support for primary/secondary process
>   net/zxdh: add support for GENEVE TSO and Rx outer checksum
> 
>  doc/guides/rel_notes/release_26_03.rst |   7 +
>  drivers/net/zxdh/zxdh_common.c         |  75 ++------
>  drivers/net/zxdh/zxdh_common.h         |   2 +-
>  drivers/net/zxdh/zxdh_ethdev.c         | 239 +++++++++++++++++--------
>  drivers/net/zxdh/zxdh_ethdev.h         |  23 ++-
>  drivers/net/zxdh/zxdh_ethdev_ops.c     | 165 ++++++++++++++---
>  drivers/net/zxdh/zxdh_ethdev_ops.h     |  15 ++
>  drivers/net/zxdh/zxdh_msg.c            |  52 ++++++
>  drivers/net/zxdh/zxdh_msg.h            |  24 ++-
>  drivers/net/zxdh/zxdh_pci.h            |   2 +
>  drivers/net/zxdh/zxdh_queue.c          | 137 +++++++++++---
>  drivers/net/zxdh/zxdh_queue.h          |  12 +-
>  drivers/net/zxdh/zxdh_rxtx.c           | 117 +++++++++---
>  drivers/net/zxdh/zxdh_tables.h         |  12 +-
>  14 files changed, 655 insertions(+), 227 deletions(-)
> 

This looks much better, there a couple of AI review complaints about
commit message style. I will address those during the merge.

For example, the AI generated commit message for the second patch
would be:

net/zxdh: optimize queue resource allocation with hardware locking

This patch refactors the queue resource allocation mechanism to improve
performance and reduce complexity by moving the allocation operation
inside the hardware lock critical section.

Key changes:
- Remove the complex zxdh_common_table_write() function and its
  associated messaging infrastructure
- Simplify zxdh_datach_set() to directly write PCIe ID to hardware
  registers using BAR0 memory mapping
- Move queue resource allocation (zxdh_datach_set) inside the hardware
  lock critical section in zxdh_get_available_channel()
- Remove the separate zxdh_datach_set() call from zxdh_reconfig_queues()
  since it's now handled during channel allocation
- Add new PCIe ID address and size definitions for direct register access

This optimization eliminates the overhead of complex message passing for
queue setup operations while ensuring atomic allocation of queue resources
under hardware lock protection.

Signed-off-by: Junlong Wang <wang.junlong1@zte.com.cn>

  parent reply	other threads:[~2026-01-31 18:12 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-20  6:15 [PATCH v1 0/5] net/zxdh: add support some new features Junlong Wang
2025-12-20  6:15 ` [PATCH v1 1/5] net/zxdh: add support for modifying queue depth Junlong Wang
2025-12-31 19:15   ` Stephen Hemminger
2025-12-20  6:15 ` [PATCH v1 2/5] net/zxdh: optimize alloc queue resources Junlong Wang
2025-12-20  6:15 ` [PATCH v1 3/5] net/zxdh: add support set link speed get autoneg stats Junlong Wang
2025-12-21 16:57   ` Stephen Hemminger
2025-12-31 19:23   ` Stephen Hemminger
2025-12-31 19:24   ` Stephen Hemminger
2025-12-31 19:27   ` Stephen Hemminger
2025-12-20  6:15 ` [PATCH v1 4/5] net/zxdh: add support primary/secondary process Junlong Wang
2025-12-21 17:00   ` Stephen Hemminger
2025-12-23 10:55   ` Junlong Wang
2025-12-23 17:54   ` Stephen Hemminger
2025-12-20  6:15 ` [PATCH v1 5/5] net/zxdh: add support GENEVE TSO and Rx outer UDP chksum Junlong Wang
2026-01-05  7:16 ` [PATCH v2 0/5] net/zxdh: add support some new features Junlong Wang
2026-01-05  7:16   ` [PATCH v2 1/5] net/zxdh: add support for modifying queue depth Junlong Wang
2026-01-05  7:16   ` [PATCH v2 2/5] net/zxdh: optimize alloc queue resources Junlong Wang
2026-01-05  7:16   ` [PATCH v2 3/5] net/zxdh: add support set link speed get autoneg stats Junlong Wang
2026-01-05  7:16   ` [PATCH v2 4/5] net/zxdh: add support primary/secondary process Junlong Wang
2026-01-05  7:16   ` [PATCH v2 5/5] net/zxdh: add support GENEVE TSO and Rx outer chksum Junlong Wang
2026-01-12  0:32   ` [PATCH v2 0/5] net/zxdh: add support some new features Stephen Hemminger
2026-01-31  6:22   ` [PATCH v3 " Junlong Wang
2026-01-31  6:22     ` [PATCH v3 1/5] net/zxdh: add support for modifying queue depth Junlong Wang
2026-01-31  6:22     ` [PATCH v3 2/5] net/zxdh: optimize alloc queue resources Junlong Wang
2026-01-31  6:22     ` [PATCH v3 3/5] net/zxdh: add link speed set and autoneg get support Junlong Wang
2026-01-31  6:22     ` [PATCH v3 4/5] net/zxdh: add support for primary/secondary process Junlong Wang
2026-01-31  6:22     ` [PATCH v3 5/5] net/zxdh: add support for GENEVE TSO and Rx outer checksum Junlong Wang
2026-03-18  2:21       ` [v3, " Junlong Wang
2026-03-18 10:16         ` Thomas Monjalon
2026-03-18 10:54           ` Thomas Monjalon
2026-03-19  5:53       ` Junlong Wang
2026-03-19  9:27         ` Thomas Monjalon
2026-01-31 18:11     ` Stephen Hemminger [this message]
2026-02-02 19:45     ` [PATCH v3 0/5] net/zxdh: add support some new features Stephen Hemminger
2026-02-03  1:43     ` [v3,0/5] " Junlong Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260131101157.4700d0e9@phoenix.local \
    --to=stephen@networkplumber.org \
    --cc=dev@dpdk.org \
    --cc=wang.junlong1@zte.com.cn \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox